need help about flir bx320?

[cq-317]

I bought a BX320 manufactured 2006 from ebay. According to the seller, the machine was only used a few times, and the machine was really new. The machine lost battery.

when i receiving the machine,i used the 12V power to power on the machine. but the machine could not be turned on, the green led is on and the screen is always black. But press and hold the power button, the green led will be off, and then the green led will be on again after turned on, but the screen does not respond.

then I made a serial cable followed the instructions and connected to the computer, I found that the machine started, but stuck somewhere, and then restarted indefinitely.

 I can enter the boot menu In the serial port , I can execute the BOOT2 command (start boot2 without application) through the boot menu , this will start the system, I can connect to the device through the network in the system. There is no reaction on the screen during this process. When I execute the GO command (start application) in the boot menu, the program will be stuck and enter an infinite restart.This looks like the application can't start.

Can someone help me? What is wrong with this bx320?

[Fraser]

Cq-317,

Sorry to hear about your cameras problem.

I am not a software expert and have only worked on the E series hardware on which your BX320 is based.

The lack of any LCD display is of concern and could be caused by the firmware looping before the display splash screen is generated. Such a loop error could be RAM or ROM related. As you are able to at least talk to the microprocessor and activate a command, a RAM failure seems unlikely to me. That leaves a possible firmware corruption as a possibility. The camera is now 12 years old so corruption of the flash RAM is possible. Another possibility is a hardware failure due to age or abuse (fluid ingress or dropping).

It would be worth seeing if everything looks OK inside the camera. Look for fluid ingress as this camera is most definitely not water resistant ! Look for green corrosion around the LCD screen and on the PCB's. Make sure all ribbon connectors are OK and correctly seated.

Re assemble the camera and check its operation. If it still will not complete the boot sequence you are left with considering trying to update the Firmware. You cannot just clone the firmware from another working E series camera as unique calibration data is involved. I am uncertain how the E series firmware update was carried out. It may need a fully booted and happy OS before starting the update. Alternatively it may use the common simultaneous button press routine to invoke the firmware update boot routine ? I would have to check my records. Not sure if FLIR still offer a firmware image but I will check.

Fraser

[Fraser]

I just checked on firmware availability for the old E series cameras. Sadly no such firmware is available from FLIR and the camera is not "field updatable"

There is no easy way to be certain what has gone wrong with your camera. I can only suggest careful visual inspection of the cameras internal parts to see if it has been dropped or water has entered its case. Repairing these cameras is very challenging without the schematics and I do it as a hobby and only when I have the camera in front of me. Remote diagnostics can be a very challenging experience for all concerned ! Sorry I cannot help more.

If there is an option to get your money back on this faulty camera, I recommend you pursue such.

Fraser

[Fraser]

In case it helps with the camera disassembly, here is my teardown of the very similar FLIR E2 camera:

https://www.eevblog.com/forum/thermal-imaging/flir-thermal-camera-for-teardown-my-latest-patient-the-e2/msg669563/#msg669563

Fraser

[cq-317]

Cq-317,

Sorry to hear about your cameras problem.

I am not a software expert and have only worked on the E series hardware on which your BX320 is based.

The lack of any LCD display is of concern and could be caused by the firmware looping before the display splash screen is generated. Such a loop error could be RAM or ROM related. As you are able to at least talk to the microprocessor and activate a command, a RAM failure seems unlikely to me. That leaves a possible firmware corruption as a possibility. The camera is now 12 years old so corruption of the flash RAM is possible. Another possibility is a hardware failure due to age or abuse (fluid ingress or dropping).

It would be worth seeing if everything looks OK inside the camera. Look for fluid ingress as this camera is most definitely not water resistant ! Look for green corrosion around the LCD screen and on the PCB's. Make sure all ribbon connectors are OK and correctly seated.

Re assemble the camera and check its operation. If it still will not complete the boot sequence you are left with considering trying to update the Firmware. You cannot just clone the firmware from another working E series camera as unique calibration data is involved. I am uncertain how the E series firmware update was carried out. It may need a fully booted and happy OS before starting the update. Alternatively it may use the common simultaneous button press routine to invoke the firmware update boot routine ? I would have to check my records. Not sure if FLIR still offer a firmware image but I will check.

Fraser

Thank you very much for your patience.

teardown this camera is very simple, all pcbs looks like very new, without any traces of water, including green corrosion.

I also found an interesting thing. If only 12V power is plugged in, the serial port will not print any data. but If me insert the battery, the serial port will print the data immediately.then the system is constantly restarting. Is there any possibility to see the startup log?

It is really a pity that for this almost new BX320, here are some teardown pictures for everyone to enjoy.

[Fraser]

The diagnostics port is that long edge connector on the largest PCB. It will likely have a serial port on it that outputs the engineering data as the camera boots. This usually contains the camera ID, all module serial numbers and the boot log in real time.

As you say, the camera is in lovely condition. It reminds me of a 'new' Agema PM570 I purchased as faulty. I could not care to leave it faulty so reverse engineered the who,e cameras hardware in order to find the fault. Turned out to be a HALT condition on the MC68340 Processor caused by a faulty Hex buffer that served a reset line on a sub system PCB ! A 50 Cent IC !

Fraser

[cq-317]

Yes, I re-examined the long edge connector on the largest PCB. I found that the 9th pin is connected to the RX pin of the external PS2 interface, and the 10th pin is connected to the TX pin of the external PS2 interface. . So the serial port in  the long edge connector is  same as and the  external  PS2 serial port.

The external PS2 interface contains a 232 serial port and a USB port. When the battery insert, the data will be printed on the 232 serial port, but only a few information in the boot,and system will restart again and again.

Code: [Select]

Mact1 Boot1: 1.2.12, build 1 28-Oct-2004 (Release)
Intel flash 16Mb, blocksize 128kb, 128 blocks detected
Press "b <enter>" to enter boot mode
BOOT:Starting program at 0xFC044004

Mainboard revision: MACT4 05
MATRIX appl: 1.7.28, build 5 06-Jul-2006 (Release)


Code: [Select]

boot> go
Calculating checksum for main program...
Mainboard revision: MACT4 05C044004
MATRIX Boot2: 2.7.4, build 2 29-Nov-2004 (Release)
àMATRIX appl: 1.7.28, build 5 06-Jul-2006 (Release)


Code: [Select]

Mact1 Boot1: 1.2.12, build 1 28-Oct-2004 (Release)
Intel flash 16Mb, blocksize 128kb, 128 blocks detected
Press "b <enter>" to enter boot mode
boot> help
Avaliable commands are:
GO           - Start main program
BOOT2        - Start boot2 without application
LOAD         - Load S-Record file to RAM
CLEAR        - Read RAM data from FLASH
PROGRAM      - Write RAM data to FLASH
VERIFY       - Compare RAM and FLASH data
ECHO         - Switch ON echo
NOECHO       - Switch OFF echo
BAUDRATE <b> - Set baudrate <b>
LOCK [N[-N]] - Lock selected flash sectors
UNLOCK       - Unlock all flash sectors
SHOWLOCK     - Show locked flash sectors
boot> boot2
Mainboard revision: MACT4 05C044004
MATRIX Boot2: 2.7.4, build 2 29-Nov-2004 (Release)
à
Copyright FLIR Systems AB, 2000-2004

OK>help


ATE0Q0V1       ATH            arp            ate0q0v1       baudrate
boot           bufstat        cat            cd             checkdisk
cleartime      clrscr         cmp            cp             date
du             echo           file           filetest       flash
flirppp        format         getfblock      getpri         head
help           i2cread        i2cwrite       ifconfig       kill
load           loadpic        ls             mainver        memstat
mf             mkdir          mount          mr             mv
mw             netstat        osstat         passwd         picload
ping           popd           power          ppp            pppaddr
pushd          putfblock      pwd            pwv            reclaim
restart        resume         rm             rmdir          route
setenv         setpri         setsize        sleep          suspend
sync           tail           tasklist       taskstk        tasktime
temptest       termecho       touch          umount         usbrd
usbwr          verify         vfmstat
OK>
When the serial port prints the boot information, pressing b+enter will enter the boot menu. Then the boot menu does not have a lot of useful information. Executing the boot2 command through the boot menu will enter the linux system. At this time, the computer will recognize the USB device to a network port, and get the ip 192.168.65.66, through telnet,computer can also connected to 192.168.65.65(BX320). most command in the help can be run.Everything shows that the linux system is working, just the application can't start.

[Fraser]

I am no software expert but I thought boo loops were often the result of the firmware 'asking' a software module for an answer, not getting it, and repeating the question again and again until the answer is received ? The failure to provide the expected answer to the query can be software or hardware based. If a hardware sub system is failing to respond, the program may just be set to repeat the query of the module until it responds. The alternative is a critical error HALT state where the program does not receive the expected response so either forks to a fault report branch or enters a HALT state. the last command or query generated on the data and address buss's usually leads you to which area of the camera is failing to respond to a query.

Fraser

[cq-317]

Yes, this kind of failure is not easy to repair because there is no schematic and any log.The only thing that can be done is to check the solder and the connectors, but this is not easy to check because most of the components are BGA packages, we can't see the pins directly.it's too hard.

Thank you for your reply, I am giving up the repair.

[railrun]

Have you already checked all power rails?

[cq-317]

I have checked some of the power rails and it looks normal. but i can't judge whether it is normal, because there is no schematic. I have given up, thank you for your help.